Need more help with FFmpeg video conversion tool

As explained in my previous post found ( here ) I am struggling with FFmpeg documentations to find parameters suitable to do VR type conversion where it is showing progress .

Now the new challenge is that I came to realize that conversion on my Intel I7 is taking roughly around 8 hours to convert a video which judging by the volume of videos I plan to convert might take me months of dedication to achieve now based on my experience with normal videos , hardware acceleration is usually the solution to cut the conversion time dramatically , after researching for hours I realize that using Nvidia Cuda acceleration has a lot of rant and hassles , so next option in line was Intel QSV , now the problem is I found so many weird syntaxes to apply such acceleration during encoding , now I need an expert help , what is the most optimized parameter to add for just using QSV during encoding ? no codec change , no anything to cause extra burden or overheads .

nIGHTmAYOR said:

Now the new challenge is that I came to realize that conversion on my Intel I7 is taking

VR, yes, but... do you have any example of the original, along with the final result you want to achieve?

I ask this because just the term VR doesn't tell me much. A VR video file can come in diff forms, such as (1) one video absolutely stereoscopic (two videos in one, being quite large to preserve quality); (2) one video also absolutely stereoscopic with slight diff angles that allow 3D perception; (3) one big video in full spherical form, some call it VR yes, but some call it immersive video; and other options that I don't have in mind right now, like proprietary formats out of the box.

I can elaborate from here, but not knowing the specifics this could be wasted effort. Also, it depends on what's the final output, like, the capabilities of the player, hardware, etc. Depending the case, you could perhaps use conversion tools like VirtualDub or Movavi converter (or others that I don't remember right now).

hroldan said:

VR, yes, but... do you have any example of the original, along with the final result you want to achieve?

I ask this because just the term VR doesn't tell me much. A VR video file can come in diff forms, such as (1) one video absolutely stereoscopic (two videos in one, being quite large to preserve quality); (2) one video also absolutely stereoscopic with slight diff angles that allow 3D perception; (3) one big video in full spherical form, some call it VR yes, but some call it immersive video; and other options that I don't have in mind right now, like proprietary formats out of the box.

I can elaborate from here, but not knowing the specifics this could be wasted effort. Also, it depends on what's the final output, like, the capabilities of the player, hardware, etc. Depending the case, you could perhaps use conversion tools like VirtualDub or Movavi converter (or others that I don't remember right now).

I may start a tutorial post to list my findings about VR conversion at some point , I always do that . also the listed tools you mention can hardly touch the geometry of VR which in that case is what is VR conversion is all about .

Like I mention in the OP , all I need is the parameters for the command line of FFMpeg to make it run in QSV mode so let us not deviate the post .

nIGHTmAYOR said:

so let us not deviate the post .

wow... asked for examples to achieve better understanding, and because you are not getting traction here. Don't mind my post then.

Added: should have said that I developed in the past my own VR viewer? when shooting at least spherical photo and videos? made my point clear before asking. Good luck there.

hroldan said:

wow... asked for examples to achieve better understanding, and because you are not getting traction here. Don't mind my post then.

Added: should have said that I developed in the past my own VR viewer? when shooting at least spherical photo and videos? made my point clear before asking. Good luck there.

I do not want to people to think its a post about vr particularly because nobody knows much about vr and probably will end up in side convos like the one we are having , all i wanned was to know is the syntax for using QSV with FFMpeg , the vr part of my question was just a hint that i need it because it takes FFMpeg an absurd time that i need to cut it by any margin .